 Discrete Fourier transform (general)

See also: Fourier transform on finite groups
This article is about the discrete Fourier transform (DFT) over any field (including finite fields), commonly called a numbertheoretic transform (NTT) in the case of finite fields. For specific information on the discrete Fourier transform over the complex numbers, see discrete Fourier transform.
Contents
Definition
Let F be any field, let be an integer, and let α be a primitive nth root of unity. We assume α lies in F.
The discrete Fourier transform maps an ntuple of elements of F to another ntuple of elements of F according to the following formula:
By convention, the tuple is said to be in the time domain and the index j is called time. The tuple is said to be in the frequency domain and the index k is called frequency. The tuple is also called the spectrum of . This terminology derives from the applications of Fourier transforms in signal processing.
Inverse
The inverse of the discrete Fourier transform is given as:
Proof: first, note that whenever β is an nth root of unity, then β^{n} − 1 = 0. If moreover , then , and therefore
Substituting (1) into the righthandside of (2), we get
This is exactly equal to v_{j}, because when (by (3) with β = α^{j' − j}), and when j' = j.
Matrix formulation
Since the discrete Fourier transform is a linear operator, it can be described by matrix multiplication. In matrix notation, the discrete Fourier transform is expressed as follows:
The matrix for this transformation is called the DFT matrix.
Similarly, the matrix notation for the inverse Fourier transform is
Polynomial formulation^{[1]}
Sometimes it is convenient to identify an ntuple with a formal polynomial
By writing out the summation in the definition of the discrete Fourier transform (1), we obtain:
This means that f_{k} is just the value of the polynomial p_{v}(x) for x = α^{k}, i.e.,
The Fourier transform can therefore be seen to relate the coefficients and the values of a polynomial: the coefficients are in the timedomain, and the values are in the frequency domain. Here, of course, it is important that the polynomial is evaluated at the nth roots of unity, which are exactly the powers of α.
Similarly, the definition of the inverse Fourier transform (2) can be written:
With
this means that
We can summarize this as follows: if the values of p(x) are the coefficients of q(x), then the values of q(x) are the coefficients of p(x), up to a scalar factor and reordering.
Special cases
Complex numbers
If is the field of complex numbers, then the nth roots of unity can be visualized as points on the unit circle of the complex plane. In this case, one usually takes
which yields the usual formula for the complex discrete Fourier transform:
Over the complex numbers, it is often customary to normalize the formulas for the DFT and inverse DFT by using the scalar factor in both formulas, rather than 1 in the formula for the DFT and in the formula for the inverse DFT. With this normalization, the DFT matrix is then unitary. Note that does not make sense in an arbitrary field.
Finite fields
If F = GF(q) is a finite field, where q is a prime power, then the existence of a primitive nth root automatically implies that n divides q − 1 (because the multiplicative order of each element must divide the size of the multiplicative group of F, which is q − 1). This in particular ensures that is invertible, so that the notation in (2) makes sense.
An application of the discrete Fourier transform over GF(q) is the reduction of ReedSolomon codes to BCH codes in coding theory.
Numbertheoretic transform
The socalled numbertheoretic transform (NTT) is obtained by specializing the discrete Fourier transform to , the integers modulo a prime p. In this case, primitive nth roots of unity exists whenever n divides p − 1, so we have p = ξn + 1 for a positive integer ξ. Specifically, let ω be a primitive (p − 1)st root of unity, then an nth root of unity α can be found by letting α = ω^{ξ}.
The numbertheoretic transform can further be generalized to operate on elements of the ring , even when the modulus m is not prime. Various special cases of the number theoretic transform such as the Fermat Number Transform or Mersenne Number Transform use a composite modulus (see e.g. Schönhage–Strassen algorithm).
For the number theoretic transform to work in a useful way when the modulus is composite (for the inverse algorithm, convolution etc. to work), it suffices that the modulus is chosen so that a primitive root of order n exists (where n is the transform length), and such that the multiplicative inverse of n exists. Note that if α is a primitive root of order n, then α is automatically invertible with inverse α^{n − 1}.
Properties
Most of the important attributes of the complex DFT, including the inverse transform, the convolution theorem, and most fast Fourier transform (FFT) algorithms, depend only on the property that the kernel of the transform is a primitive root of unity. These properties also hold, with identical proofs, over arbitrary fields. This analogy can be formalized by the field with one element, considering any field with a primitive nth root of unity as an algebra over the extension field
In particular, the applicability of O(nlog n) fast Fourier transform algorithms to compute the NTT, combined with the convolution theorem, mean that the numbertheoretic transform gives an efficient way to compute exact convolutions of integer sequences. While the complex DFT can perform the same task, it is susceptible to roundoff error in finiteprecision floating point arithmetic; the NTT has no roundoff because it deals purely with fixedsize integers that can be exactly represented.
Fast algorithms
For the implementation of a "fast" algorithm (similar to how FFT computes the DFT), it is often desirable that the transform length is also highly composite, e.g. a power of two. However, there are specialized fast Fourier transform algorithms for finite fields, such as Wang and Zhu's algorithm^{[2]}, which are efficient regardless of whether the transform length factors.
See also
References
 ^ R. Lidl and G. Pilz. Applied Abstract Algebra, 2nd edition. Wiley, 1999, pp. 217219.
 ^ Yao Wang and Xuelong Zhu, "A fast algorithm for the Fourier transform over finite fields and its VLSI implementation", IEEE Journal on Selected Areas in Communications 6(3)572577, 1988
External links
Categories:
Wikimedia Foundation. 2010.